What happens when duplicates are detected across different backends?

There is a default precedence defined for each backend. This should be defined so that the backend with highest precedence is the backend that should contain the most precise information usually.

When deduplicating, the post from the backend with the highest precedence is kept and missing info is taken from the duplicate posts (precedence is used so that in case of conflicts in a field, the data from the backend with highest precedence is used). This post contains as much data as possible, and includes references to all the other “duplicate” posts. These latter duplicate posts are then simply marked as such and never shown anymore.

All origins are kept in a urls field in the remaining post.

Flatisfy seems to be stuck fetching posts

Fetching posts can be a long process, depending on your criterias. Run the import command with -v argument to get a more verbose output and check things are indeed happening. If fetching the flats is still too long, try to set max_entries in your config to limit the number of posts fetched.

Docker image does not start the webserver at first start?

When you launch the Docker image, it first updates WebOOB and fetches the housing posts matching your criterias. The webserver is only started once this is done. As fetching housing posts can take a bit of time (up to 10 minutes), the webserver will not be available right away.

Once everything is ready, you should see a log message in the console running the Docker image, confirming you that webserver is up and running.